Exit to userspace when emulating an atomic guest access if the CMPXCHG on
the userspace address faults. Emulating the access as a write and thus
likely treating it as emulated MMIO is wrong, as KVM has already
confirmed there is a valid, writable memslot.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220202004945.2540433-6-seanjc@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The issue occurs when the devfreq cooling device uses the EM power model
and the get_real_power() callback is provided by the driver.
The EM power table is sorted ascending，can't index the table by cooling
device state，so convert cooling state to performance state by
dfc->max_state - dfc->capped_state.
Fixes: 615510fe13bd ("thermal: devfreq_cooling: remove old power model and use EM") Cc: 5.11+ <stable@vger.kernel.org> # 5.11+ Signed-off-by: Ye Zhang <ye.zhang@rock-chips.com> Reviewed-by: Dhruva Gole <d-gole@ti.com> Reviewed-by: Lukasz Luba <lukasz.luba@arm.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Lukasz Luba <lukasz.luba@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Sven reports an infinite loop in __alloc_pages_slowpath() for costly order
__GFP_RETRY_MAYFAIL allocations that are also GFP_NOIO. Such combination
can happen in a suspend/resume context where a GFP_KERNEL allocation can
have __GFP_IO masked out via gfp_allowed_mask.
Quoting Sven:
1. try to do a "costly" allocation (order > PAGE_ALLOC_COSTLY_ORDER)
with __GFP_RETRY_MAYFAIL set.
2. page alloc's __alloc_pages_slowpath tries to get a page from the
freelist. This fails because there is nothing free of that costly
order.
3. page alloc tries to reclaim by calling __alloc_pages_direct_reclaim,
which bails out because a zone is ready to be compacted; it pretends
to have made a single page of progress.
4. page alloc tries to compact, but this always bails out early because
__GFP_IO is not set (it's not passed by the snd allocator, and even
if it were, we are suspending so the __GFP_IO flag would be cleared
anyway).
5. page alloc believes reclaim progress was made (because of the
pretense in item 3) and so it checks whether it should retry
compaction. The compaction retry logic thinks it should try again,
because:
a) reclaim is needed because of the early bail-out in item 4
b) a zonelist is suitable for compaction
6. goto 2. indefinite stall.
(end quote)
The immediate root cause is confusing the COMPACT_SKIPPED returned from
__alloc_pages_direct_compact() (step 4) due to lack of __GFP_IO to be
indicating a lack of order-0 pages, and in step 5 evaluating that in
should_compact_retry() as a reason to retry, before incrementing and
limiting the number of retries. There are however other places that
wrongly assume that compaction can happen while we lack __GFP_IO.
To fix this, introduce gfp_compaction_allowed() to abstract the __GFP_IO
evaluation and switch the open-coded test in try_to_compact_pages() to use
it.
Also use the new helper in:
- compaction_ready(), which will make reclaim not bail out in step 3, so
there's at least one attempt to actually reclaim, even if chances are
small for a costly order
- in_reclaim_compaction() which will make should_continue_reclaim()
return false and we don't over-reclaim unnecessarily
- in __alloc_pages_slowpath() to set a local variable can_compact,
which is then used to avoid retrying reclaim/compaction for costly
allocations (step 5) if we can't compact and also to skip the early
compaction attempt that we do in some cases
Link: https://lkml.kernel.org/r/20240221114357.13655-2-vbabka@suse.cz Fixes: 3250845d0526 ("Revert "mm, oom: prevent premature OOM killer invocation for high order request"") Signed-off-by: Vlastimil Babka <vbabka@suse.cz> Reported-by: Sven van Ashbrook <svenva@chromium.org> Closes: https://lore.kernel.org/all/CAG-rBihs_xMKb3wrMO1%2B-%2Bp4fowP9oy1pa_OTkfxBzPUVOZF%2Bg@mail.gmail.com/ Tested-by: Karthikeyan Ramasubramanian <kramasub@chromium.org> Cc: Brian Geffon <bgeffon@google.com> Cc: Curtis Malainey <cujomalainey@chromium.org> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Mel Gorman <mgorman@techsingularity.net> Cc: Michal Hocko <mhocko@kernel.org> Cc: Takashi Iwai <tiwai@suse.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Since d794734c9bbf was also marked for -stable, let's back it out before
causing more damage.
Note that due to another upstream change the revert was not 100% automatic:
0a845e0f6348 mm/treewide: replace pud_large() with pud_leaf()
Signed-off-by: Ingo Molnar <mingo@kernel.org> Cc: <stable@vger.kernel.org> Cc: Russ Anderson <rja@hpe.com> Cc: Steve Wahl <steve.wahl@hpe.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Link: https://lore.kernel.org/all/3a1b9909-45ac-4f97-ad68-d16ef1ce99db@pavinjoseph.com/ Fixes: d794734c9bbf ("x86/mm/ident_map: Use gbpages only where full GB page should be mapped.") Signed-off-by: Steve Wahl <steve.wahl@hpe.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
io_uring: ensure '0' is returned on file registration success
A previous backport mistakenly removed code that cleared 'ret' to zero,
as the SCM logging was performed. Fix up the return value so we don't
return an errant error on fixed file registration.
Fixes: d909d381c315 ("io_uring: drop any code related to SCM_RIGHTS") Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Make the region inside the rwsem_write_trylock non preemptible.
We observe RT task is hogging CPU when trying to acquire rwsem lock
which was acquired by a kworker task but before the rwsem owner was set.
Here is the scenario:
1. CFS task (affined to a particular CPU) takes rwsem lock.
2. CFS task gets preempted by a RT task before setting owner.
3. RT task (FIFO) is trying to acquire the lock, but spinning until
RT throttling happens for the lock as the lock was taken by CFS task.
This patch attempts to fix the above issue by disabling preemption
until owner is set for the lock. While at it also fix the issues
at the places where rwsem_{set,clear}_owner() are called.
This also adds lockdep annotation of preemption disable in
rwsem_{set,clear}_owner() on Peter Z. suggestion.
Signed-off-by: Gokul krishna Krishnakumar <quic_gokukris@quicinc.com> Signed-off-by: Mukesh Ojha <quic_mojha@quicinc.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Waiman Long <longman@redhat.com> Link: https://lore.kernel.org/r/1662661467-24203-1-git-send-email-quic_mojha@quicinc.com Cc: Aaro Koskinen <aaro.koskinen@iki.fi> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* The code modified by this patch is guarded by IS_ERR(trans_private),
where trans_private is assigned as per the previous point in this analysis.
The only implementation of get_mr that I could locate is rds_ib_get_mr()
which can return an ERR_PTR if the conn (4th) argument is NULL.
* ret is set to PTR_ERR(trans_private).
rds_ib_get_mr can return ERR_PTR(-ENODEV) if the conn (4th) argument is NULL.
Thus ret may be -ENODEV in which case the code in question will execute.
Conclusion:
* cp may be NULL at the point where this patch adds a check;
this patch does seem to address a possible bug
Fixes: c055fc00c07b ("net/rds: fix WARNING in rds_conn_connect_if_down") Cc: stable@vger.kernel.org # v4.19+ Signed-off-by: Mahmoud Adam <mngyadam@amazon.com> Reviewed-by: Simon Horman <horms@kernel.org> Link: https://lore.kernel.org/r/20240326153132.55580-1-mngyadam@amazon.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Notice that skb_mark_for_recycle() is introduced later than fixes tag in
commit 6a5bcd84e886 ("page_pool: Allow drivers to hint on SKB recycling").
It is believed that fixes tag were missing a call to page_pool_release_page()
between v5.9 to v5.14, after which is should have used skb_mark_for_recycle().
Since v6.6 the call page_pool_release_page() were removed (in
commit 535b9c61bdef ("net: page_pool: hide page_pool_release_page()")
and remaining callers converted (in commit 6bfef2ec0172 ("Merge branch
'net-page_pool-remove-page_pool_release_page'")).
This leak became visible in v6.8 via commit dba1b8a7ab68 ("mm/page_pool: catch
page_pool memory leaks").
Cc: stable@vger.kernel.org Fixes: 6c5aa6fc4def ("xen networking: add basic XDP support for xen-netfront") Reported-by: Leonidas Spyropoulos <artafinde@archlinux.com> Link: https://bugzilla.kernel.org/show_bug.cgi?id=218654 Reported-by: Arthur Borsboom <arthurborsboom@gmail.com> Signed-off-by: Jesper Dangaard Brouer <hawk@kernel.org> Link: https://lore.kernel.org/r/171154167446.2671062.9127105384591237363.stgit@firesoul Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
struct hci_dev members conn_info_max_age, conn_info_min_age,
le_conn_max_interval, le_conn_min_interval, le_adv_max_interval,
and le_adv_min_interval can be modified from the HCI core code, as well
through debugfs.
The debugfs implementation, that's only available to privileged users,
will check for boundaries, making sure that the minimum value being set
is strictly above the maximum value that already exists, and vice-versa.
However, as both minimum and maximum values can be changed concurrently
to us modifying them, we need to make sure that the value we check is
the value we end up using.
For example, with ->conn_info_max_age set to 10, conn_info_min_age_set()
gets called from vfs handlers to set conn_info_min_age to 8.
In conn_info_min_age_set(), this goes through:
if (val == 0 || val > hdev->conn_info_max_age)
return -EINVAL;
Concurrently, conn_info_max_age_set() gets called to set to set the
conn_info_max_age to 7:
if (val == 0 || val > hdev->conn_info_max_age)
return -EINVAL;
That check will also pass because we used the old value (10) for
conn_info_max_age.
After those checks that both passed, the struct hci_dev access
is mutex-locked, disabling concurrent access, but that does not matter
because the invalid value checks both passed, and we'll end up with
conn_info_min_age = 8 and conn_info_max_age = 7
To fix this problem, we need to lock the structure access before so the
check and assignment are not interrupted.
This fix was originally devised by the BassCheck[1] team, and
considered the problem to be an atomicity one. This isn't the case as
there aren't any concerns about the variable changing while we check it,
but rather after we check it parallel to another change.
This patch fixes CVE-2024-24858 and CVE-2024-24857.
We have a BT headset (Lenovo Thinkplus XT99), the pairing and
connecting has no problem, once this headset is paired, bluez will
remember this device and will auto re-connect it whenever the device
is powered on. The auto re-connecting works well with Windows and
Android, but with Linux, it always fails. Through debugging, we found
at the rfcomm connection stage, the bluetooth stack reports
"Connection refused - security block (0x0003)".
For this device, the re-connecting negotiation process is different
from other BT headsets, it sends the Link_KEY_REQUEST command before
the CONNECT_REQUEST completes, and it doesn't send ENCRYPT_CHANGE
command during the negotiation. When the device sends the "connect
complete" to hci, the ev->encr_mode is 1.
So here in the conn_complete_evt(), if ev->encr_mode is 1, link type
is ACL and HCI_CONN_ENCRYPT is not set, we set HCI_CONN_ENCRYPT to
this conn, and update conn->enc_key_size accordingly.
After this change, this BT headset could re-connect with Linux
successfully. This is the btmon log after applying the patch, after
receiving the "Connect Complete" with "Encryption: Enabled", will send
the command to read encryption key size:
> HCI Event: Connect Request (0x04) plen 10
Address: 8C:3C:AA:D8:11:67 (OUI 8C-3C-AA)
Class: 0x240404
Major class: Audio/Video (headset, speaker, stereo, video, vcr)
Minor class: Wearable Headset Device
Rendering (Printing, Speaker)
Audio (Speaker, Microphone, Headset)
Link type: ACL (0x01)
...
> HCI Event: Link Key Request (0x17) plen 6
Address: 8C:3C:AA:D8:11:67 (OUI 8C-3C-AA)
< HCI Command: Link Key Request Reply (0x01|0x000b) plen 22
Address: 8C:3C:AA:D8:11:67 (OUI 8C-3C-AA)
Link key: ${32-hex-digits-key}
...
> HCI Event: Connect Complete (0x03) plen 11
Status: Success (0x00)
Handle: 256
Address: 8C:3C:AA:D8:11:67 (OUI 8C-3C-AA)
Link type: ACL (0x01)
Encryption: Enabled (0x01)
< HCI Command: Read Encryption Key... (0x05|0x0008) plen 2
Handle: 256
< ACL Data TX: Handle 256 flags 0x00 dlen 10
L2CAP: Information Request (0x0a) ident 1 len 2
Type: Extended features supported (0x0002)
> HCI Event: Command Complete (0x0e) plen 7
Read Encryption Key Size (0x05|0x0008) ncmd 1
Status: Success (0x00)
Handle: 256
Key size: 16
Cc: stable@vger.kernel.org Link: https://github.com/bluez/bluez/issues/704 Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de> Reviewed-by: Luiz Augusto von Dentz <luiz.dentz@gmail.com> Signed-off-by: Hui Wang <hui.wang@canonical.com> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Several Qualcomm Bluetooth controllers lack persistent storage for the
device address and instead one can be provided by the boot firmware
using the 'local-bd-address' devicetree property.
The Bluetooth bindings clearly states that the address should be
specified in little-endian order, but due to a long-standing bug in the
Qualcomm driver which reversed the address some boot firmware has been
providing the address in big-endian order instead.
The boot firmware in SC7180 Trogdor Chromebooks is known to be affected
so mark the 'local-bd-address' property as broken to maintain backwards
compatibility with older firmware when fixing the underlying driver bug.
Note that ChromeOS always updates the kernel and devicetree in lockstep
so that there is no need to handle backwards compatibility with older
devicetrees.
Fixes: 7ec3e67307f8 ("arm64: dts: qcom: sc7180-trogdor: add initial trogdor and lazor dt") Cc: stable@vger.kernel.org # 5.10 Cc: Rob Clark <robdclark@chromium.org> Reviewed-by: Douglas Anderson <dianders@chromium.org> Signed-off-by: Johan Hovold <johan+linaro@kernel.org> Acked-by: Bjorn Andersson <andersson@kernel.org> Reviewed-by: Bjorn Andersson <andersson@kernel.org> Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add CPUID_LNX_5 to track cpufeatures' word 21, and add the appropriate
compile-time assert in KVM to prevent direct lookups on the features in
CPUID_LNX_5. KVM uses X86_FEATURE_* flags to manage guest CPUID, and so
must translate features that are scattered by Linux from the Linux-defined
bit to the hardware-defined bit, i.e. should never try to directly access
scattered features in guest CPUID.
Opportunistically add NR_CPUID_WORDS to enum cpuid_leafs, along with a
compile-time assert in KVM's CPUID infrastructure to ensure that future
additions update cpuid_leafs along with NCAPINTS.
No functional change intended.
Fixes: 7f274e609f3d ("x86/cpufeatures: Add new word for scattered features") Cc: Sandipan Das <sandipan.das@amd.com> Signed-off-by: Sean Christopherson <seanjc@google.com> Acked-by: Dave Hansen <dave.hansen@linux.intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On some boards with this chip version the BIOS is buggy and misses
to reset the PHY page selector. This results in the PHY ID read
accessing registers on a different page, returning a more or
less random value. Fix this by resetting the page selector first.
Depending on the value of CONFIG_HZ, clang complains about a pointless
comparison:
drivers/md/dm-integrity.c:4085:12: error: result of comparison of
constant 42949672950 with expression of type
'unsigned int' is always false
[-Werror,-Wtautological-constant-out-of-range-compare]
if (val >= (uint64_t)UINT_MAX * 1000 / HZ) {
As the check remains useful for other configurations, shut up the
warning by adding a second type cast to uint64_t.
This patch re-introduces protection against the size of access to stack
memory being negative; the access size can appear negative as a result
of overflowing its signed int representation. This should not actually
happen, as there are other protections along the way, but we should
protect against it anyway. One code path was missing such protections
(fixed in the previous patch in the series), causing out-of-bounds array
accesses in check_stack_range_initialized(). This patch causes the
verification of a program with such a non-sensical access size to fail.
This check used to exist in a more indirect way, but was inadvertendly
removed in a833a17aeac7.
The mlxbf_gige driver encounters a NULL pointer exception in
mlxbf_gige_open() when kdump is enabled. The sequence to reproduce
the exception is as follows:
a) enable kdump
b) trigger kdump via "echo c > /proc/sysrq-trigger"
c) kdump kernel executes
d) kdump kernel loads mlxbf_gige module
e) the mlxbf_gige module runs its open() as the
the "oob_net0" interface is brought up
f) mlxbf_gige module will experience an exception
during its open(), something like:
The exception happens because there is a pending RX interrupt before the
call to request_irq(RX IRQ) executes. Then, the RX IRQ handler fires
immediately after this request_irq() completes. The RX IRQ handler runs
"napi_schedule()" before NAPI is fully initialized via "netif_napi_add()"
and "napi_enable()", both which happen later in the open() logic.
The logic in mlxbf_gige_open() must fully initialize NAPI before any calls
to request_irq() execute.
We had various syzbot reports about tcp timers firing after
the corresponding netns has been dismantled.
Fortunately Josef Bacik could trigger the issue more often,
and could test a patch I wrote two years ago.
When TCP sockets are closed, we call inet_csk_clear_xmit_timers()
to 'stop' the timers.
inet_csk_clear_xmit_timers() can be called from any context,
including when socket lock is held.
This is the reason it uses sk_stop_timer(), aka del_timer().
This means that ongoing timers might finish much later.
For user sockets, this is fine because each running timer
holds a reference on the socket, and the user socket holds
a reference on the netns.
For kernel sockets, we risk that the netns is freed before
timer can complete, because kernel sockets do not hold
reference on the netns.
This patch adds inet_csk_clear_xmit_timers_sync() function
that using sk_stop_timer_sync() to make sure all timers
are terminated before the kernel socket is released.
Modules using kernel sockets close them in their netns exit()
handler.
Also add sock_not_owned_by_me() helper to get LOCKDEP
support : inet_csk_clear_xmit_timers_sync() must not be called
while socket lock is held.
It is very possible we can revert in the future commit 3a58f13a881e ("net: rds: acquire refcount on TCP sockets")
which attempted to solve the issue in rds only.
(net/smc/af_smc.c and net/mptcp/subflow.c have similar code)
We probably can remove the check_net() tests from
tcp_out_of_resources() and __tcp_close() in the future.
Reported-by: Josef Bacik <josef@toxicpanda.com> Closes: https://lore.kernel.org/netdev/20240314210740.GA2823176@perftesting/ Fixes: 26abe14379f8 ("net: Modify sk_alloc to not reference count the netns of kernel sockets.") Fixes: 8a68173691f0 ("net: sk_clone_lock() should only do get_net() if the parent is not a kernel socket") Link: https://lore.kernel.org/bpf/CANn89i+484ffqb93aQm1N-tjxxvb3WDKX0EbD7318RwRgsatjw@mail.gmail.com/ Signed-off-by: Eric Dumazet <edumazet@google.com> Tested-by: Josef Bacik <josef@toxicpanda.com> Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Link: https://lore.kernel.org/r/20240322135732.1535772-1-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
The IO subsystem expects a driver to retry a ccw_device_start, when the
subsequent interrupt response block (irb) contains a deferred
condition code 1.
Symptoms before this commit:
On the read channel we always trigger the next read anyhow, so no
different behaviour here.
On the write channel we may experience timeout errors, because the
expected reply will never be received without the retry.
Other callers of qeth_send_control_data() may wrongly assume that the ccw
was successful, which may cause problems later.
Note that since
commit 2297791c92d0 ("s390/cio: dont unregister subchannel from child-drivers")
and
commit 5ef1dc40ffa6 ("s390/cio: fix invalid -EBUSY on ccw_device_start")
deferred CC1s are much more likely to occur. See the commit message of the
latter for more background information.
Fixes: 2297791c92d0 ("s390/cio: dont unregister subchannel from child-drivers") Signed-off-by: Alexandra Winter <wintera@linux.ibm.com> Co-developed-by: Thorsten Winkler <twinkler@linux.ibm.com> Signed-off-by: Thorsten Winkler <twinkler@linux.ibm.com> Reviewed-by: Peter Oberparleiter <oberpar@linux.ibm.com> Link: https://lore.kernel.org/r/20240321115337.3564694-1-wintera@linux.ibm.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Change kzalloc() flags used in ixgbe_ipsec_vf_add_sa() to GFP_ATOMIC, to
avoid sleeping in IRQ context.
Dan Carpenter, with the help of Smatch, has found following issue:
The patch eda0333ac293: "ixgbe: add VF IPsec management" from Aug 13,
2018 (linux-next), leads to the following Smatch static checker
warning: drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c:917 ixgbe_ipsec_vf_add_sa()
warn: sleeping in IRQ context
The call tree that Smatch is worried about is:
ixgbe_msix_other() <- IRQ handler
-> ixgbe_msg_task()
-> ixgbe_rcv_msg_from_vf()
-> ixgbe_ipsec_vf_add_sa()
Fixes: eda0333ac293 ("ixgbe: add VF IPsec management") Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Link: https://lore.kernel.org/intel-wired-lan/db31a0b0-4d9f-4e6b-aed8-88266eb5665c@moroto.mountain Reviewed-by: Michal Kubiak <michal.kubiak@intel.com> Signed-off-by: Przemek Kitszel <przemyslaw.kitszel@intel.com> Reviewed-by: Shannon Nelson <shannon.nelson@amd.com> Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel) Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
syzbot reported the following uninit-value access issue [1][2]:
nci_rx_work() parses and processes received packet. When the payload
length is zero, each message type handler reads uninitialized payload
and KMSAN detects this issue. The receipt of a packet with a zero-size
payload is considered unexpected, and therefore, such packets should be
silently discarded.
This patch resolved this issue by checking payload size before calling
each message type handler codes.
Fixes: 6a2968aaf50c ("NFC: basic NCI protocol implementation") Reported-and-tested-by: syzbot+7ea9413ea6749baf5574@syzkaller.appspotmail.com Reported-and-tested-by: syzbot+29b5ca705d2e0f4a44d2@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=7ea9413ea6749baf5574 [1] Closes: https://syzkaller.appspot.com/bug?extid=29b5ca705d2e0f4a44d2 [2] Signed-off-by: Ryosuke Yasuoka <ryasuoka@redhat.com> Reviewed-by: Jeremy Cline <jeremy@jcline.org> Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
In the scenario of entering hibernation with udisk in the system, if the
udisk was gone or resume fail in the thaw phase of hibernation. Its state
will be set to NOTATTACHED. At this point, usb_hub_wq was already freezed
and can't not handle disconnect event. Next, in the poweroff phase of
hibernation, SYNCHRONIZE_CACHE SCSI command will be sent to this udisk
when poweroff this scsi device, which will cause uas_submit_urbs to be
called to submit URB for sense/data/cmd pipe. However, these URBs will
submit fail as device was set to NOTATTACHED state. Then, uas_submit_urbs
will return a value SCSI_MLQUEUE_DEVICE_BUSY to the caller. That will lead
the SCSI layer go into an ugly loop and system fail to go into hibernation.
On the other hand, when we specially check for -ENODEV in function
uas_queuecommand_lck, returning DID_ERROR to SCSI layer will cause device
poweroff fail and system shutdown instead of entering hibernation.
To fix this issue, let uas_submit_urbs to return original generic error
when submitting URB failed. At the same time, we need to translate -ENODEV
to DID_NOT_CONNECT for the SCSI layer.
Suggested-by: Oliver Neukum <oneukum@suse.com> Cc: stable@vger.kernel.org Signed-off-by: Weitao Wang <WeitaoWang-oc@zhaoxin.com> Link: https://lore.kernel.org/r/20240306180814.4897-1-WeitaoWang-oc@zhaoxin.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
Set scsi_host_template.cmd_size instead of using the SCSI pointer for
storing driver-private data. Change the type of the argument of
uas_add_work() from struct uas_cmd_info * into struct scsi_cmnd * because
it is easier to convert a SCSI command pointer into a uas_cmd_info pointer
than the other way around.
This patch prepares for removal of the SCSI pointer from struct scsi_cmnd.
Link: https://lore.kernel.org/r/20220218195117.25689-46-bvanassche@acm.org Cc: linux-usb@vger.kernel.org Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com> Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Acked-by: Oliver Neukum <oneukum@suse.com> Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Stable-dep-of: cd5432c71235 ("USB: UAS: return ENODEV when submit urbs fail with device not attached") Signed-off-by: Sasha Levin <sashal@kernel.org>
Conditional statements are faster than indirect calls. Hence call
scsi_done() directly.
Link: https://lore.kernel.org/r/20211007204618.2196847-10-bvanassche@acm.org Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Stable-dep-of: cd5432c71235 ("USB: UAS: return ENODEV when submit urbs fail with device not attached") Signed-off-by: Sasha Levin <sashal@kernel.org>
Among the attribute file callback routines in
drivers/usb/core/sysfs.c, the interface_authorized_store() function is
the only one which acquires a device lock on an ancestor device: It
calls usb_deauthorize_interface(), which locks the interface's parent
USB device.
The will lead to deadlock if another process already owns that lock
and tries to remove the interface, whether through a configuration
change or because the device has been disconnected. As part of the
removal procedure, device_del() waits for all ongoing sysfs attribute
callbacks to complete. But usb_deauthorize_interface() can't complete
until the device lock has been released, and the lock won't be
released until the removal has finished.
The mechanism provided by sysfs to prevent this kind of deadlock is
to use the sysfs_break_active_protection() function, which tells sysfs
not to wait for the attribute callback.
Reported-and-tested by: Yue Sun <samsun1006219@gmail.com>
Reported by: xingwei lee <xrivendell7@gmail.com>
The wqe is of type lpfc_wqe128. It should be memset with the same type.
Fixes: 6c621a2229b0 ("scsi: lpfc: Separate NVMET RQ buffer posting from IO resources SGL/iocbq/context") Signed-off-by: Muhammad Usama Anjum <usama.anjum@collabora.com> Link: https://lore.kernel.org/r/20240304090649.833953-1-usama.anjum@collabora.com Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Reviewed-by: Justin Tee <justintee8345@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 5459c0b70467 ("PCI/DPC: Quirk PIO log size for certain Intel Root
Ports") added quirks for Tiger and Alder Lake Root Ports but missed that
the same issue exists also in the previous generation, Ice Lake.
Apply the quirk for Ice Lake Root Ports as well. This prevents kernel
complaints like:
DPC: RP PIO log size 0 is invalid
and also enables the DPC driver to dump the RP PIO Log registers when DPC
is triggered.
[bhelgaas: add dmesg warning and RP PIO Log dump info] Closes: https://bugzilla.kernel.org/show_bug.cgi?id=209943 Link: https://lore.kernel.org/r/20230511121905.73949-1-mika.westerberg@linux.intel.com Reported-by: Mark Blakeney <mark.blakeney@bullet-systems.net> Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Unlike Intel's Enhanced IBRS feature, AMD's Automatic IBRS does not
provide protection to processes running at CPL3/user mode, see section
"Extended Feature Enable Register (EFER)" in the APM v2 at
https://bugzilla.kernel.org/attachment.cgi?id=304652
Explicitly enable STIBP to protect against cross-thread CPL3
branch target injections on systems with Automatic IBRS enabled.
Also update the relevant documentation.
Fixes: e7862eda309e ("x86/cpu: Support AMD Automatic IBRS") Reported-by: Tom Lendacky <thomas.lendacky@amd.com> Signed-off-by: Kim Phillips <kim.phillips@amd.com> Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de> Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20230720194727.67022-1-kim.phillips@amd.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently when PCI error is detected, I/O is aborted manually through the
ABORT IOCB mechanism which is not guaranteed to succeed.
Instead, wait for the OS or system to notify driver to wind down I/O
through the pci_error_handlers api. Set eeh_busy flag to pause all traffic
and wait for I/O to drain.
Upon driver unload, purge_mbox flag is set and the heartbeat monitor thread
detects this flag and does not send the mailbox command down to FW with a
debug message "Error detected: purge[1] eeh[0] cmd=0x0, Exiting". This
being not a real error, change the debug message.
The system was under memory stress where driver was not able to allocate an
SRB to carry out error recovery of cable pull. The failure to flush causes
upper layer to start modifying scsi_cmnd. When the system frees up some
memory, the subsequent cable pull trigger another command flush. At this
point the driver access a null pointer when attempting to DMA unmap the
SGL.
Add a check to make sure commands are flush back on session tear down to
prevent the null pointer access.
Changing of [FCP|NVME] prefer flag in flash has no effect on driver. For
device that supports both FCP + NVMe over the same connection, driver
continues to connect to this device using the previous successful login
mode.
On completion of flash update, adapter will be reset. Driver will
reset the prefer flag based on setting from flash.
Disk failed to rediscover after chip reset error injection. The chip reset
happens at the time when a PLOGI is being sent. This causes a flag to be
left on which blocks the retry. Clear the blocking flag.
Currently IOCBs are allowed to push through while chip reset could be in
progress. During chip reset the outstanding_cmds array is cleared
twice. Once when any command on this array is returned as failed and
secondly when the array is initialize to zero. If a command is inserted on
to the array between these intervals, then the command will be lost. Check
for chip reset before sending IOCB.
Check the UCSI_CCI_RESET_COMPLETE complete flag before starting
another reset. Use a UCSI_SET_NOTIFICATION_ENABLE command to clear
the flag if it is set.
Signed-off-by: Christian A. Ehrhardt <lk@c--e.de> Cc: stable <stable@kernel.org> Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-QRD Link: https://lore.kernel.org/r/20240320073927.1641788-6-lk@c--e.de Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If a command completes the OPM must send an ack. This applies
to unsupported commands, too.
Send the required ACK for unsupported commands.
Signed-off-by: Christian A. Ehrhardt <lk@c--e.de> Cc: stable <stable@kernel.org> Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-QRD Link: https://lore.kernel.org/r/20240320073927.1641788-4-lk@c--e.de Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Added functionality to exit from L1 state by device initiation
using remote wakeup signaling, in case when function driver queuing
request while core in L1 state.
Added exiting from the clock gating mode on USB Reset Detect interrupt
if core in the clock gating mode.
Added new condition to check core in clock gating mode or no.
Fixed ISOC completion flow in DDMA mode. Added isoc
descriptor actual length value and update urb's start_frame
value.
Fixed initialization of ISOC DMA descriptors flow.
Fixes: 56f5b1cff22a ("staging: Core files for the DWC2 driver") Fixes: 20f2eb9c4cf8 ("staging: dwc2: add microframe scheduler from downstream Pi kernel") Fixes: c17b337c1ea4 ("usb: dwc2: host: program descriptor for next frame") Fixes: dc4c76e7b22c ("staging: HCD descriptor DMA support for the DWC2 driver") Fixes: 762d3a1a9cd7 ("usb: dwc2: host: process all completed urbs") CC: stable@vger.kernel.org Signed-off-by: Minas Harutyunyan <Minas.Harutyunyan@synopsys.com> Link: https://lore.kernel.org/r/a8b1e1711cc6cabfb45d92ede12e35445c66f06c.1708944698.git.Minas.Harutyunyan@synopsys.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Create hub_get() and hub_put() routines to encapsulate the kref_get()
and kref_put() calls in hub.c. The new routines will be used by the
next patch in this series.
The m.u.component_create.pid field is for debugging and in the mainline
kernel it's not used anything. However, it still needs to be set to
something to prevent disclosing uninitialized stack data. Set it to
zero.
Fixes: 7b3ad5abf027 ("staging: Import the BCM2835 MMAL-based V4L2 camera driver.") Cc: stable <stable@kernel.org> Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Link: https://lore.kernel.org/r/2d972847-9ebd-481b-b6f9-af390f5aabd3@moroto.mountain Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
gcc-14 warns about this strncpy() that results in a non-terminated
string for an overflow:
In file included from include/linux/string.h:369,
from drivers/staging/vc04_services/vchiq-mmal/mmal-vchiq.c:20:
In function 'strncpy',
inlined from 'create_component' at drivers/staging/vc04_services/vchiq-mmal/mmal-vchiq.c:940:2:
include/linux/fortify-string.h:108:33: error: '__builtin_strncpy' specified bound 128 equals destination size [-Werror=stringop-truncation]
Change it to strscpy_pad(), which produces a properly terminated and
zero-padded string.
Commit fc663711b944 ("scsi: core: Remove the /proc/scsi/${proc_name}
directory earlier") fixed a bug related to modules loading/unloading, by
adding a call to scsi_proc_hostdir_rm() on scsi_remove_host(). But that led
to a potential duplicate call to the hostdir_rm() routine, since it's also
called from scsi_host_dev_release(). That triggered a regression report,
which was then fixed by commit be03df3d4bfe ("scsi: core: Fix a procfs host
directory removal regression"). The fix just dropped the hostdir_rm() call
from dev_release().
But it happens that this proc directory is created on scsi_host_alloc(),
and that function "pairs" with scsi_host_dev_release(), while
scsi_remove_host() pairs with scsi_add_host(). In other words, it seems the
reason for removing the proc directory on dev_release() was meant to cover
cases in which a SCSI host structure was allocated, but the call to
scsi_add_host() didn't happen. And that pattern happens to exist in some
error paths, for example.
Syzkaller causes that by using USB raw gadget device, error'ing on
usb-storage driver, at usb_stor_probe2(). By checking that path, we can see
that the BadDevice label leads to a scsi_host_put() after a SCSI host
allocation, but there's no call to scsi_add_host() in such path. That leads
to messages like this in dmesg (and a leak of the SCSI host proc
structure):
usb-storage 4-1:87.51: USB Mass Storage device detected
proc_dir_entry 'scsi/usb-storage' already registered
WARNING: CPU: 1 PID: 3519 at fs/proc/generic.c:377 proc_register+0x347/0x4e0 fs/proc/generic.c:376
The proper fix seems to still call scsi_proc_hostdir_rm() on dev_release(),
but guard that with the state check for SHOST_CREATED; there is even a
comment in scsi_host_dev_release() detailing that: such conditional is
meant for cases where the SCSI host was allocated but there was no calls to
{add,remove}_host(), like the usb-storage case.
This is what we propose here and with that, the error path of usb-storage
does not trigger the warning anymore.
Reported-by: syzbot+c645abf505ed21f931b5@syzkaller.appspotmail.com Fixes: be03df3d4bfe ("scsi: core: Fix a procfs host directory removal regression") Cc: stable@vger.kernel.org Cc: Bart Van Assche <bvanassche@acm.org> Cc: John Garry <john.g.garry@oracle.com> Cc: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com> Signed-off-by: Guilherme G. Piccoli <gpiccoli@igalia.com> Link: https://lore.kernel.org/r/20240313113006.2834799-1-gpiccoli@igalia.com Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The dreamcastcard->timer could schedule the spu_dma_work and the
spu_dma_work could also arm the dreamcastcard->timer.
When the snd_pcm_substream is closing, the aica_channel will be
deallocated. But it could still be dereferenced in the worker
thread. The reason is that del_timer() will return directly
regardless of whether the timer handler is running or not and
the worker could be rescheduled in the timer handler. As a result,
the UAF bug will happen. The racy situation is shown below:
In order to mitigate this bug and other possible corner cases,
call mod_timer() conditionally in run_spu_dma(), then implement
PCM sync_stop op to cancel both the timer and worker. The sync_stop
op will be called from PCM core appropriately when needed.
Fixes: 198de43d758c ("[ALSA] Add ALSA support for the SEGA Dreamcast PCM device") Suggested-by: Takashi Iwai <tiwai@suse.de> Signed-off-by: Duoming Zhou <duoming@zju.edu.cn>
Message-ID: <20240326094238.95442-1-duoming@zju.edu.cn> Signed-off-by: Takashi Iwai <tiwai@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[Why]
The original picture aspect ratio in mode struct may have chance be
overwritten with wrong aspect ratio data in create_stream_for_sink().
It will create a different VIC output and cause HDMI compliance test
failed.
[How]
Preserve the original picture aspect ratio data during create the
stream.
Cc: Mario Limonciello <mario.limonciello@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: stable@vger.kernel.org Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Tom Chung <chiahsuan.chung@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
struct drm_display_mode embeds a list head, so overwriting
the full struct with another one will corrupt the list
(if the destination mode is on a list). Use drm_mode_copy()
instead which explicitly preserves the list head of
the destination mode.
Even if we know the destination mode is not on any list
using drm_mode_copy() seems decent as it sets a good
example. Bad examples of not using it might eventually
get copied into code where preserving the list head
actually matters.
Obviously one case not covered here is when the mode
itself is embedded in a larger structure and the whole
structure is copied. But if we are careful when copying
into modes embedded in structures I think we can be a
little more reassured that bogus list heads haven't been
propagated in.
@is_mode_copy@
@@
drm_mode_copy(...)
{
...
}
@depends on !is_mode_copy@
struct drm_display_mode *mode;
expression E, S;
@@
(
- *mode = E
+ drm_mode_copy(mode, &E)
|
- memcpy(mode, E, S)
+ drm_mode_copy(mode, E)
)
@depends on !is_mode_copy@
struct drm_display_mode mode;
expression E;
@@
(
- mode = E
+ drm_mode_copy(&mode, &E)
|
- memcpy(&mode, E, S)
+ drm_mode_copy(&mode, E)
)
Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Harry Wentland <harry.wentland@amd.com> Cc: Leo Li <sunpeng.li@amd.com> Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Cc: amd-gfx@lists.freedesktop.org Reviewed-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Stable-dep-of: 79f3e38f60e5 ("drm/amd/display: Preserve original aspect ratio in create stream") Signed-off-by: Sasha Levin <sashal@kernel.org>
wdm_read() cannot race with itself. However, in
service_outstanding_interrupt() it can race with the
workqueue, which can be triggered by error handling.
Hence we need to make sure that the WDM_RESPONDING
flag is not just only set but tested.
Originally, with strict in order execution, we could complete execution
only when the queue was empty. Preempt-to-busy allows replacement of an
active request that may complete before the preemption is processed by
HW. If that happens, the request is retired from the queue, but the
queue_priority_hint remains set, preventing direct submission until
after the next CS interrupt is processed.
This preempt-to-busy race can be triggered by the heartbeat, which will
also act as the power-management barrier and upon completion allow us to
idle the HW. We may process the completion of the heartbeat, and begin
parking the engine before the CS event that restores the
queue_priority_hint, causing us to fail the assertion that it is MIN.
It is necessary for the queue_priority_hint to be lower than the next
request submission upon waking up, as we rely on the hint to decide when
to kick the tasklet to submit that first request.
Fixes: 22b7a426bbe1 ("drm/i915/execlists: Preempt-to-busy") Closes: https://gitlab.freedesktop.org/drm/intel/issues/10154 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Janusz Krzysztofik <janusz.krzysztofik@linux.intel.com> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: <stable@vger.kernel.org> # v5.4+ Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com> Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240318135906.716055-2-janusz.krzysztofik@linux.intel.com
(cherry picked from commit 98850e96cf811dc2d0a7d0af491caff9f5d49c1e) Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 4d0c8d0aef63 ("mmc: core: Use mrq.sbc in close-ended ffu") assigns
prev_idata = idatas[i - 1], but doesn't check that the iterator i is
greater than zero. Let's fix this by adding a check.
Commit 4d0c8d0aef63 ("mmc: core: Use mrq.sbc in close-ended ffu") adds
flags uint to struct mmc_blk_ioc_data, but it does not get initialized for
RPMB ioctls which now fails.
Let's fix this by always initializing the struct and flags to zero.
Fixes: 4d0c8d0aef63 ("mmc: core: Use mrq.sbc in close-ended ffu") Closes: https://bugzilla.kernel.org/show_bug.cgi?id=218587 Link: https://lore.kernel.org/all/20231129092535.3278-1-avri.altman@wdc.com/ Cc: stable@vger.kernel.org Signed-off-by: Mikko Rapeli <mikko.rapeli@linaro.org> Reviewed-by: Avri Altman <avri.altman@wdc.com> Acked-by: Adrian Hunter <adrian.hunter@intel.com> Tested-by: Francesco Dolcini <francesco.dolcini@toradex.com> Link: https://lore.kernel.org/r/20240313133744.2405325-1-mikko.rapeli@linaro.org Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
After the linked LLVM change, the build fails with
CONFIG_LD_ORPHAN_WARN_LEVEL="error", which happens with allmodconfig:
ld.lld: error: vmlinux.a(init/main.o):(.hexagon.attributes) is being placed in '.hexagon.attributes'
Handle the attributes section in a similar manner as arm and riscv by
adding it after the primary ELF_DETAILS grouping in vmlinux.lds.S, which
fixes the error.
In NOMMU kernel the value of linux_binprm::p is the offset inside the
temporary program arguments array maintained in separate pages in the
linux_binprm::page. linux_binprm::exec being a copy of linux_binprm::p
thus must be adjusted when that array is copied to the user stack.
Without that adjustment the value passed by the NOMMU kernel to the ELF
program in the AT_EXECFN entry of the aux array doesn't make any sense
and it may break programs that try to access memory pointed to by that
entry.
Adjust linux_binprm::exec before the successful return from the
transfer_args_to_stack().
When moving a station out of a VLAN and deleting the VLAN afterwards, the
fast_rx entry still holds a pointer to the VLAN's netdev, which can cause
use-after-free bugs. Fix this by immediately calling ieee80211_check_fast_rx
after the VLAN change.
Cc: stable@vger.kernel.org Reported-by: ranygh@riseup.net Signed-off-by: Felix Fietkau <nbd@nbd.name> Link: https://msgid.link/20240316074336.40442-1-nbd@nbd.name Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Zi Yan [Wed, 6 Mar 2024 15:51:57 +0000 (10:51 -0500)]
mm/migrate: set swap entry values of THP tail pages properly.
The tail pages in a THP can have swap entry information stored in their
private field. When migrating to a new page, all tail pages of the new
page need to update ->private to avoid future data corruption.
This fix is stable-only, since after commit 07e09c483cbe ("mm/huge_memory:
work on folio->swap instead of page->private when splitting folio"),
subpages of a swapcached THP no longer requires the maintenance.
Adding THPs to the swapcache was introduced in commit 38d8b4e6bdc87 ("mm, THP, swap: delay splitting THP during swap out"),
where each subpage of a THP added to the swapcache had its own swapcache
entry and required the ->private field to point to the correct swapcache
entry. Later, when THP migration functionality was implemented in commit 616b8371539a6 ("mm: thp: enable thp migration in generic path"),
it initially did not handle the subpages of swapcached THPs, failing to
update their ->private fields or replace the subpage pointers in the
swapcache. Subsequently, commit e71769ae5260 ("mm: enable thp migration
for shmem thp") addressed the swapcache update aspect. This patch fixes
the update of subpage ->private fields.
Closes: https://lore.kernel.org/linux-mm/1707814102-22682-1-git-send-email-quic_charante@quicinc.com/ Fixes: 616b8371539a ("mm: thp: enable thp migration in generic path") Signed-off-by: Zi Yan <ziy@nvidia.com> Acked-by: David Hildenbrand <david@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The SC16IS7XX IC supports a burst mode to access the FIFOs where the
initial register address is sent ($00), followed by all the FIFO data
without having to resend the register address each time. In this mode, the
IC doesn't increment the register address for each R/W byte.
The regmap_raw_read() and regmap_raw_write() are functions which can
perform IO over multiple registers. They are currently used to read/write
from/to the FIFO, and although they operate correctly in this burst mode on
the SPI bus, they would corrupt the regmap cache if it was not disabled
manually. The reason is that when the R/W size is more than 1 byte, these
functions assume that the register address is incremented and handle the
cache accordingly.
Convert FIFO R/W functions to use the regmap _noinc_ versions in order to
remove the manual cache control which was a workaround when using the
_raw_ versions. FIFO registers are properly declared as volatile so
cache will not be used/updated for FIFO accesses.
The eventfd_ctx trigger pointer of the vfio_fsl_mc_irq object is
initially NULL and may become NULL if the user sets the trigger
eventfd to -1. The interrupt handler itself is guaranteed that
trigger is always valid between request_irq() and free_irq(), but
the loopback testing mechanisms to invoke the handler function
need to test the trigger. The triggering and setting ioctl paths
both make use of igate and are therefore mutually exclusive.
The vfio-fsl-mc driver does not make use of irqfds, nor does it
support any sort of masking operations, therefore unlike vfio-pci
and vfio-platform, the flow can remain essentially unchanged.
Cc: Diana Craciun <diana.craciun@oss.nxp.com> Cc: <stable@vger.kernel.org> Fixes: cc0ee20bd969 ("vfio/fsl-mc: trigger an interrupt via eventfd") Reviewed-by: Kevin Tian <kevin.tian@intel.com> Reviewed-by: Eric Auger <eric.auger@redhat.com> Link: https://lore.kernel.org/r/20240308230557.805580-8-alex.williamson@redhat.com Signed-off-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The vfio-platform SET_IRQS ioctl currently allows loopback triggering of
an interrupt before a signaling eventfd has been configured by the user,
which thereby allows a NULL pointer dereference.
Rather than register the IRQ relative to a valid trigger, register all
IRQs in a disabled state in the device open path. This allows mask
operations on the IRQ to nest within the overall enable state governed
by a valid eventfd signal. This decouples @masked, protected by the
@locked spinlock from @trigger, protected via the @igate mutex.
In doing so, it's guaranteed that changes to @trigger cannot race the
IRQ handlers because the IRQ handler is synchronously disabled before
modifying the trigger, and loopback triggering of the IRQ via ioctl is
safe due to serialization with trigger changes via igate.
For compatibility, request_irq() failures are maintained to be local to
the SET_IRQS ioctl rather than a fatal error in the open device path.
This allows, for example, a userspace driver with polling mode support
to continue to work regardless of moving the request_irq() call site.
This necessarily blocks all SET_IRQS access to the failed index.
Cc: Eric Auger <eric.auger@redhat.com> Cc: <stable@vger.kernel.org> Fixes: 57f972e2b341 ("vfio/platform: trigger an interrupt via eventfd") Reviewed-by: Kevin Tian <kevin.tian@intel.com> Reviewed-by: Eric Auger <eric.auger@redhat.com> Link: https://lore.kernel.org/r/20240308230557.805580-7-alex.williamson@redhat.com Signed-off-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A vulnerability exists where the eventfd for INTx signaling can be
deconfigured, which unregisters the IRQ handler but still allows
eventfds to be signaled with a NULL context through the SET_IRQS ioctl
or through unmask irqfd if the device interrupt is pending.
Ideally this could be solved with some additional locking; the igate
mutex serializes the ioctl and config space accesses, and the interrupt
handler is unregistered relative to the trigger, but the irqfd path
runs asynchronous to those. The igate mutex cannot be acquired from the
atomic context of the eventfd wake function. Disabling the irqfd
relative to the eventfd registration is potentially incompatible with
existing userspace.
As a result, the solution implemented here moves configuration of the
INTx interrupt handler to track the lifetime of the INTx context object
and irq_type configuration, rather than registration of a particular
trigger eventfd. Synchronization is added between the ioctl path and
eventfd_signal() wrapper such that the eventfd trigger can be
dynamically updated relative to in-flight interrupts or irqfd callbacks.
In order to synchronize changes that can affect the thread callback,
introduce an interface to force a flush of the inject workqueue. The
irqfd pointer is only valid under spinlock, but the workqueue cannot
be flushed under spinlock. Therefore the flush work for the irqfd is
queued under spinlock. The vfio_irqfd_cleanup_wq workqueue is re-used
for queuing this work such that flushing the workqueue is also ordered
relative to shutdown.
Reviewed-by: Kevin Tian <kevin.tian@intel.com> Reviewed-by: Reinette Chatre <reinette.chatre@intel.com> Reviewed-by: Eric Auger <eric.auger@redhat.com> Link: https://lore.kernel.org/r/20240308230557.805580-4-alex.williamson@redhat.com Signed-off-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Mask operations through config space changes to DisINTx may race INTx
configuration changes via ioctl. Create wrappers that add locking for
paths outside of the core interrupt code.
In particular, irq_type is updated holding igate, therefore testing
is_intx() requires holding igate. For example clearing DisINTx from
config space can otherwise race changes of the interrupt configuration.
This aligns interfaces which may trigger the INTx eventfd into two
camps, one side serialized by igate and the other only enabled while
INTx is configured. A subsequent patch introduces synchronization for
the latter flows.
Currently for devices requiring masking at the irqchip for INTx, ie.
devices without DisINTx support, the IRQ is enabled in request_irq()
and subsequently disabled as necessary to align with the masked status
flag. This presents a window where the interrupt could fire between
these events, resulting in the IRQ incrementing the disable depth twice.
This would be unrecoverable for a user since the masked flag prevents
nested enables through vfio.
Instead, invert the logic using IRQF_NO_AUTOEN such that exclusive INTx
is never auto-enabled, then unmask as required.
Cc: <stable@vger.kernel.org> Fixes: 89e1f7d4c66d ("vfio: Add PCI device driver") Reviewed-by: Kevin Tian <kevin.tian@intel.com> Reviewed-by: Eric Auger <eric.auger@redhat.com> Link: https://lore.kernel.org/r/20240308230557.805580-2-alex.williamson@redhat.com Signed-off-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The test counter 'test_cnt' should not be returned in diag.sh, e.g. what
if only the 4th test fail? Will do 'exit 4' which is 'exit ${KSFT_SKIP}',
the whole test will be marked as skipped instead of 'failed'!
arch/powerpc/lib/xor_vmx.o is built with '-msoft-float' (from the main
powerpc Makefile) and '-maltivec' (from its CFLAGS), which causes an
error when building with clang after a recent change in main:
error: option '-msoft-float' cannot be specified with '-maltivec'
make[6]: *** [scripts/Makefile.build:243: arch/powerpc/lib/xor_vmx.o] Error 1
Explicitly add '-mhard-float' before '-maltivec' in xor_vmx.o's CFLAGS
to override the previous inclusion of '-msoft-float' (as the last option
wins), which matches how other areas of the kernel use '-maltivec', such
as AMDGPU.
Work around a quirk in a few old (2011-ish) UEFI implementations, where
a call to `GetNextVariableName` with a buffer size larger than 512 bytes
will always return EFI_INVALID_PARAMETER.
There is some lore around EFI variable names being up to 1024 bytes in
size, but this has no basis in the UEFI specification, and the upper
bounds are typically platform specific, and apply to the entire variable
(name plus payload).
Given that Linux does not permit creating files with names longer than
NAME_MAX (255) bytes, 512 bytes (== 256 UTF-16 characters) is a
reasonable limit.
Cc: <stable@vger.kernel.org> # 6.1+ Signed-off-by: Tim Schumacher <timschumi@gmx.de> Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
[timschumi@gmx.de: adjusted diff for changed context and code move] Signed-off-by: Tim Schumacher <timschumi@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
perf_output_read_group may respond to IPI request of other cores and invoke
__perf_install_in_context function. As a result, hwc configuration is modified.
causing inconsistency and unexpected consequences.
Interrupts are not disabled when perf_output_read_group reads PMU counter.
In this case, IPI request may be received from other cores.
As a result, PMU configuration is modified and an error occurs when
reading PMU counter:
Commit bb4d53d66e4b ("NFSD: use (un)lock_inode instead of
fh_(un)lock for file operations") broke the NFSv3 pre/post op
attributes behaviour when doing a SETATTR rpc call by stripping out
the calls to fh_fill_pre_attrs() and fh_fill_post_attrs().
Fixes: bb4d53d66e4b ("NFSD: use (un)lock_inode instead of fh_(un)lock for file operations") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Reviewed-by: NeilBrown <neilb@suse.de>
Message-ID: <20240216012451.22725-1-trondmy@kernel.org>
[ cel: adjusted to apply to v5.15.y ] Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
It is possible for free_blocked_lock() to be called twice concurrently,
once from nfsd4_lock() and once from nfsd4_release_lockowner() calling
remove_blocked_locks(). This is why a kref was added.
It is perfectly safe for locks_delete_block() and kref_put() to be
called in parallel as they use locking or atomicity respectively as
protection. However locks_release_private() has no locking. It is
safe for it to be called twice sequentially, but not concurrently.
This patch moves that call from free_blocked_lock() where it could race
with itself, to free_nbl() where it cannot. This will slightly delay
the freeing of private info or release of the owner - but not by much.
It is arguably more natural for this freeing to happen in free_nbl()
where the structure itself is freed.
This bug was found by code inspection - it has not been seen in practice.
Fixes: 47446d74f170 ("nfsd4: add refcount for nfsd4_blocked_lock") Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
A recent change to check_for_locks() changed it to take ->flc_lock while
holding ->fi_lock. This creates a lock inversion (reported by lockdep)
because there is a case where ->fi_lock is taken while holding
->flc_lock.
->flc_lock is held across ->fl_lmops callbacks, and
nfsd_break_deleg_cb() is one of those and does take ->fi_lock. However
it doesn't need to.
Prior to v4.17-rc1~110^2~22 ("nfsd: create a separate lease for each
delegation") nfsd_break_deleg_cb() would walk the ->fi_delegations list
and so needed the lock. Since then it doesn't walk the list and doesn't
need the lock.
Two actions are performed under the lock. One is to call
nfsd_break_one_deleg which calls nfsd4_run_cb(). These doesn't act on
the nfs4_file at all, so don't need the lock.
The other is to set ->fi_had_conflict which is in the nfs4_file.
This field is only ever set here (except when initialised to false)
so there is no possible problem will multiple threads racing when
setting it.
The field is tested twice in nfs4_set_delegation(). The first test does
not hold a lock and is documented as an opportunistic optimisation, so
it doesn't impose any need to hold ->fi_lock while setting
->fi_had_conflict.
The second test in nfs4_set_delegation() *is* make under ->fi_lock, so
removing the locking when ->fi_had_conflict is set could make a change.
The change could only be interesting if ->fi_had_conflict tested as
false even though nfsd_break_one_deleg() ran before ->fi_lock was
unlocked. i.e. while hash_delegation_locked() was running.
As hash_delegation_lock() doesn't interact in any way with nfs4_run_cb()
there can be no importance to this interaction.
So this patch removes the locking from nfsd_break_one_deleg() and moves
the final test on ->fi_had_conflict out of the locked region to make it
clear that locking isn't important to the test. It is still tested
*after* vfs_setlease() has succeeded. This might be significant and as
vfs_setlease() takes ->flc_lock, and nfsd_break_one_deleg() is called
under ->flc_lock this "after" is a true ordering provided by a spinlock.
Fixes: edcf9725150e ("nfsd: fix RELEASE_LOCKOWNER") Signed-off-by: NeilBrown <neilb@suse.de> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
The test on so_count in nfsd4_release_lockowner() is nonsense and
harmful. Revert to using check_for_locks(), changing that to not sleep.
First: harmful.
As is documented in the kdoc comment for nfsd4_release_lockowner(), the
test on so_count can transiently return a false positive resulting in a
return of NFS4ERR_LOCKS_HELD when in fact no locks are held. This is
clearly a protocol violation and with the Linux NFS client it can cause
incorrect behaviour.
If RELEASE_LOCKOWNER is sent while some other thread is still
processing a LOCK request which failed because, at the time that request
was received, the given owner held a conflicting lock, then the nfsd
thread processing that LOCK request can hold a reference (conflock) to
the lock owner that causes nfsd4_release_lockowner() to return an
incorrect error.
The Linux NFS client ignores that NFS4ERR_LOCKS_HELD error because it
never sends NFS4_RELEASE_LOCKOWNER without first releasing any locks, so
it knows that the error is impossible. It assumes the lock owner was in
fact released so it feels free to use the same lock owner identifier in
some later locking request.
When it does reuse a lock owner identifier for which a previous RELEASE
failed, it will naturally use a lock_seqid of zero. However the server,
which didn't release the lock owner, will expect a larger lock_seqid and
so will respond with NFS4ERR_BAD_SEQID.
So clearly it is harmful to allow a false positive, which testing
so_count allows.
The test is nonsense because ... well... it doesn't mean anything.
so_count is the sum of three different counts.
1/ the set of states listed on so_stateids
2/ the set of active vfs locks owned by any of those states
3/ various transient counts such as for conflicting locks.
When it is tested against '2' it is clear that one of these is the
transient reference obtained by find_lockowner_str_locked(). It is not
clear what the other one is expected to be.
In practice, the count is often 2 because there is precisely one state
on so_stateids. If there were more, this would fail.
In my testing I see two circumstances when RELEASE_LOCKOWNER is called.
In one case, CLOSE is called before RELEASE_LOCKOWNER. That results in
all the lock states being removed, and so the lockowner being discarded
(it is removed when there are no more references which usually happens
when the lock state is discarded). When nfsd4_release_lockowner() finds
that the lock owner doesn't exist, it returns success.
The other case shows an so_count of '2' and precisely one state listed
in so_stateid. It appears that the Linux client uses a separate lock
owner for each file resulting in one lock state per lock owner, so this
test on '2' is safe. For another client it might not be safe.
So this patch changes check_for_locks() to use the (newish)
find_any_file_locked() so that it doesn't take a reference on the
nfs4_file and so never calls nfsd_file_put(), and so never sleeps. With
this check is it safe to restore the use of check_for_locks() rather
than testing so_count against the mysterious '2'.
Fixes: ce3c4ad7f4ce ("NFSD: Fix possible sleep during nfsd4_release_lockowner()") Signed-off-by: NeilBrown <neilb@suse.de> Reviewed-by: Jeff Layton <jlayton@kernel.org> Cc: stable@vger.kernel.org # v6.2+ Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
It's not safe to call nfsd_put once nfsd_last_thread has been called, as
that function will zero out the nn->nfsd_serv pointer.
Drop the nfsd_put helper altogether and open-code the svc_put in its
callers instead. That allows us to not be reliant on the value of that
pointer when handling an error.
Fixes: 2a501f55cd64 ("nfsd: call nfsd_last_thread() before final nfsd_put()") Reported-by: Zhi Li <yieli@redhat.com> Cc: NeilBrown <neilb@suse.de> Signed-off-by: Jeffrey Layton <jlayton@redhat.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
If write_ports_addfd or write_ports_addxprt fail, they call nfsd_put()
without calling nfsd_last_thread(). This leaves nn->nfsd_serv pointing
to a structure that has been freed.
So remove 'static' from nfsd_last_thread() and call it when the
nfsd_serv is about to be destroyed.
Fixes: ec52361df99b ("SUNRPC: stop using ->sv_nrthreads as a refcount") Signed-off-by: NeilBrown <neilb@suse.de> Reviewed-by: Jeff Layton <jlayton@kernel.org> Cc: <stable@vger.kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
This patch reverts mostly commit 40595cdc93ed ("nfs: block notification
on fs with its own ->lock") and introduces an EXPORT_OP_ASYNC_LOCK
export flag to signal that the "own ->lock" implementation supports
async lock requests. The only main user is DLM that is used by GFS2 and
OCFS2 filesystem. Those implement their own lock() implementation and
return FILE_LOCK_DEFERRED as return value. Since commit 40595cdc93ed
("nfs: block notification on fs with its own ->lock") the DLM
implementation were never updated. This patch should prepare for DLM
to set the EXPORT_OP_ASYNC_LOCK export flag and update the DLM
plock implementation regarding to it.
Acked-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Alexander Aring <aahringo@redhat.com> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
If /proc/fs/nfsd/pool_stats is open when the last nfsd thread exits, then
when the file is closed a NULL pointer is dereferenced.
This is because nfsd_pool_stats_release() assumes that the
pointer to the svc_serv cannot become NULL while a reference is held.
This used to be the case but a recent patch split nfsd_last_thread() out
from nfsd_put(), and clearing the pointer is done in nfsd_last_thread().
This is easily reproduced by running
rpc.nfsd 8 ; ( rpc.nfsd 0;true) < /proc/fs/nfsd/pool_stats
Fortunately nfsd_pool_stats_release() has easy access to the svc_serv
pointer, and so can call svc_put() on it directly.
Fixes: 9f28a971ee9f ("nfsd: separate nfsd_last_thread() from nfsd_put()") Signed-off-by: NeilBrown <neilb@suse.de> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Now that the last nfsd thread is stopped by an explicit act of calling
svc_set_num_threads() with a count of zero, we only have a limited
number of places that can happen, and don't need to call
nfsd_last_thread() in nfsd_put()
So separate that out and call it at the two places where the number of
threads is set to zero.
Move the clearing of ->nfsd_serv and the call to svc_xprt_destroy_all()
into nfsd_last_thread(), as they are really part of the same action.
nfsd_put() is now a thin wrapper around svc_put(), so make it a static
inline.
nfsd_put() cannot be called after nfsd_last_thread(), so in a couple of
places we have to use svc_put() instead.
Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Previously a thread could exit asynchronously (due to a signal) so some
care was needed to hold nfsd_mutex over the last svc_put() call. Now a
thread can only exit when svc_set_num_threads() is called, and this is
always called under nfsd_mutex. So no care is needed.
Not only is the mutex held when a thread exits now, but the svc refcount
is elevated, so the svc_put() in svc_exit_thread() will never be a final
put, so the mutex isn't even needed at this point in the code.
Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
In nfsd4_encode_fattr(), TIME_CREATE was being written out after all
other times. However, they should be written out in an order that
matches the bit flags in bmval1, which in this case are
I noticed this on a FreeBSD NFSv4.2 client, which supports creation
times. On this client, file times were weirdly permuted. With this
patch applied on the server, times looked normal on the client.
Fixes: e377a3e698fb ("nfsd: Add support for the birth time attribute") Link: https://unix.stackexchange.com/q/749605/56202 Signed-off-by: Tavian Barnes <tavianator@tavianator.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
The below-mentioned patch was intended to simplify refcounting on the
svc_serv used by locked. The goal was to only ever have a single
reference from the single thread. To that end we dropped a call to
lockd_start_svc() (except when creating thread) which would take a
reference, and dropped the svc_put(serv) that would drop that reference.
Unfortunately we didn't also remove the svc_get() from
lockd_create_svc() in the case where the svc_serv already existed.
So after the patch:
- on the first call the svc_serv was allocated and the one reference
was given to the thread, so there are no extra references
- on subsequent calls svc_get() was called so there is now an extra
reference.
This is clearly not consistent.
The inconsistency is also clear in the current code in lockd_get()
takes *two* references, one on nlmsvc_serv and one by incrementing
nlmsvc_users. This clearly does not match lockd_put().
So: drop that svc_get() from lockd_get() (which used to be in
lockd_create_svc().
The bug here is that you cannot rely on getting the same socket
from multiple calls to fget() because userspace can influence
that. This is a kind of double fetch bug.
The fix is to delete the svc_alien_sock() function and instead do
the checking inside the svc_addsock() function.
Fixes: 3064639423c4 ("nfsd: check passed socket's net matches NFSd superblock's one") Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Reviewed-by: NeilBrown <neilb@suse.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
notify_change can modify the iattr structure. In particular it can
end up setting ATTR_MODE when ATTR_KILL_SUID is already set, causing
a BUG() if the same iattr is passed to notify_change more than once.
Make a copy of the struct iattr before calling notify_change.
Reported-by: Zhi Li <yieli@redhat.com> Link: https://bugzilla.redhat.com/show_bug.cgi?id=2207969 Tested-by: Zhi Li <yieli@redhat.com> Fixes: 34b91dda7124 ("NFSD: Make nfsd4_setattr() wait before returning NFS4ERR_DELAY") Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
The following request sequence to the same file causes the NFS client and
server getting into an infinite loop with COMMIT and NFS4ERR_DELAY:
OPEN
REMOVE
WRITE
COMMIT
Problem reported by recall11, recall12, recall14, recall20, recall22,
recall40, recall42, recall48, recall50 of nfstest suite.
This patch restores the handling of race condition in nfsd_file_do_acquire
with unlink to that prior of the regression.
Fixes: ac3a2585f018 ("nfsd: rework refcounting in filecache") Signed-off-by: Dai Ngo <dai.ngo@oracle.com> Reviewed-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
When queueing a dispose list to the appropriate "freeme" lists, it
pointlessly queues the objects one at a time to an intermediate list.
Remove a few helpers and just open code a list_move to make it more
clear and efficient. Better document the resulting functions with
kerneldoc comments.
Signed-off-by: Jeff Layton <jlayton@kernel.org> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
While we were converting the nfs4_file hashtable to use the kernel's
resizable hashtable data structure, Neil Brown observed that the
list variant (rhltable) would be better for managing nfsd_file items
as well. The nfsd_file hash table will contain multiple entries for
the same inode -- these should be kept together on a list. And, it
could be possible for exotic or malicious client behavior to cause
the hash table to resize itself on every insertion.
A nice simplification is that rhltable_lookup() can return a list
that contains only nfsd_file items that match a given inode, which
enables us to eliminate specialized hash table helper functions and
use the default functions provided by the rhashtable implementation).
Since we are now storing nfsd_file items for the same inode on a
single list, that effectively reduces the number of hash entries
that have to be tracked in the hash table. The mininum bucket count
is therefore lowered.
Light testing with fstests generic/531 show no regressions.
Suggested-by: Neil Brown <neilb@suse.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
On most filesystems, there is no reason to delay reaping an nfsd_file
just because its underlying inode is still under writeback. nfsd just
relies on client activity or the local flusher threads to do writeback.
The main exception is NFS, which flushes all of its dirty data on last
close. Add a new EXPORT_OP_FLUSH_ON_CLOSE flag to allow filesystems to
signal that they do this, and only skip closing files under writeback on
such filesystems.
Also, remove a redundant NULL file pointer check in
nfsd_file_check_writeback, and clean up nfs's export op flag
definitions.
Signed-off-by: Jeff Layton <jlayton@kernel.org> Acked-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
[ cel: adjusted to apply to v5.15.y ] Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
projects / users / dwmw2 / linux.git / log